|
| Analysis of Teaching and Prompting Strategies as They Effect Acquisition of Chained Responses |
| Saturday, May 29, 2004 |
| 1:00 PM–2:20 PM |
| Back Bay C |
| Area: AUT/DDA; Domain: Applied Research |
| Chair: Myrna Libby (New England Center for Children) |
| Discussant: Richard M. Foxx (Pennsylvania State University, Harrisburg) |
| Abstract: N/a |
| |
| Acquisition of Chains Using Single vs. Multiple Teachers |
| JULIE S. WEISS (New England Center for Children), Myrna Libby (New England Center for Children), Susan N. Langer (New England Center for Children), William H. Ahearn (New England Center for Children), Richard B. Graff (New England Center for Children) |
| Abstract: Two participants diagnosed with autism each learned to put together two eight-step arbitrary Lego figures in a forward chaining sequence with most-to-least prompting. An alternating treatments design was used to compare acquisition of these multi-step tasks with either a single teacher or multiple (4) teachers running acquisition trials. Each session consisted of 1 probe trial and 10 training trials. After acquisition, generalization probes were run with two novel teachers and in a different environment. Both participants achieved independence in building the two constructs across teacher conditions. Acquisition averaged two sessions longer for the multiple teacher condition compared to the single teacher condition. In addition, both participants generalized performance across novel teachers and in a different environment for both figures. All sessions were videotaped. Procedural integrity for both the single and multiple teacher conditions was better than 90% as was IOA for of the sessions evaluated (33% of total). |
| |
| Manual Guidance vs. Teacher Completion of Chained Tasks |
| MARIA GARRETT (New England Center for Children), Myrna Libby (New England Center for Children), Julie S. Weiss (New England Center for Children), Susan N. Langer (New England Center for Children), William H. Ahearn (New England Center for Children), Richard B. Graff (New England Center for Children) |
| Abstract: An alternating treatments design was used to compare a forward chaining sequence and most-to-least prompting procedure under two conditions; during condition 1, the teacher completed the steps beyond the training step; during condition 2, the teacher manually guided the participant to complete the steps beyond the training step. Three participants diagnosed with autism participated and the dependent variable was acquisition of two eight-step arbitrary Lego figures; each session consisted of 1 probe trial and 10 training trials. Generalization probes across two novel teachers and one new setting were conducted after acquisition. Both procedures were effective in teaching the chains to all participants; however, manually guiding the participant to complete the task was, in all cases, more efficient than having the teacher complete the task. Manually guiding produced acquisition in fewer trials, with fewer errors and with no overall difference in training time. All participants generalized responding across two novel teachers and in a new environment. IOA was collected in at least 80% of sessions and averaged 95%. Procedural integrity data were taken in at least 80% of sessions and averaged 97%. |
| |
| Most-to-Least Compared to Least-to-Most Prompting in the Acquisition of Chained Task |
| STACIE L. FITCH (New England Center for Children), Myrna Libby (New England Center for Children), Julie S. Weiss (New England Center for Children), Susan N. Langer (New England Center for Children), William H. Ahearn (New England Center for Children), Richard B. Graff (New England Center for Children) |
| Abstract: Two participants diagnosed with autism were taught to put together two eight-step arbitrary Lego figures using an alternating treatments design comparing least-to-most and most-to-least prompting hierarchies. Generalization probes across two novel teachers and one new setting were conducted after acquisition. Both participants acquired both chains indicating that both prompting procedures were effective. However, most-to-least prompting produced much more efficient acquisition as measured by trials to criterion, errors and total training time. Both participants generalized across two novel teachers and a different training environment. IOA and procedural integrity were taken in 33% of sessions and averaged 90% or better. |
|
| |